An image sensor conventionally comprises a plurality of identical or similar pixels, for example arranged in an array of rows and columns. Each pixel comprises a photoelectric conversion element, for example, a photodiode, capable of generating a photocurrent representative of a received light intensity. Each pixel, further comprises a capacitive storage element capable of integrating the photocurrent generated by the photoelectric conversion element. The measurement of the illumination level received by the pixel is performed by measuring a voltage variation across the capacitive storage element between the beginning and the end of an integration phase.
To maximize the signal-to-noise ratio, it is generally desired for the light flow converted by the photoelectric con version element of the pixel during the integration period to be as high as possible, or, in other words, for the quantity of photogenerated charges stored in the capacitive storage element of the pixel during the integration period to be as high as possible. This implies for the capacitive element to have a relatively significant charge storage capacity, which goes against the usual search for pixel miniaturization, and particularly for a reduction of the pitch between pixels.